سال انتشار: ۱۳۹۰

محل انتشار: هفتمین کنفرانس ماشین بینایی و پردازش تصویر

تعداد صفحات: ۵

نویسنده(ها):

Nima Forouzanfar – Department of Electrical EngineeringSahand University of TechnologyTabriz, Iran
Hossein Ebrahimnezhad – Department of Electrical EngineeringSahand University of TechnologyTabriz, Iran
Afshin Ebrahimi – Department of Electrical EngineeringSahand University of TechnologyTabriz, Iran

چکیده:

Dynamic textures are image sequences which containspatial-varying and time varying phenomenon. There are manyexamples of visual patterns in the real-world scenes that arecalled dynamic textures. In this paper, we propose a novelanalytical approach for segmentation of dynamic textures. Thekey idea is to exploit the properties of both frame texture andmotion. In our approach, we employ linear predictive coding toextract special features from sequences of mixtures of dynamictexture.